[EM] Brief Ranked-Pairs definition

[Chris Benham]

Mike,

> Elect the candidate not beaten among all the strongest defeats that 
> don’t contradict (form any cycles with) each other. 

That definition is "brief" but isn't complete, i.e. it doesn't give 
enough information to know how to operate the method because it doesn't 
explain what "strongest" means (or exactly how we measure "strength").

As you know, more than one way has been suggested/advocated.   You like 
"Winning Votes"  (that used to be called on EM "Votes Against").  There 
used to be some people here who advocated for Margins.

I like Approval Margins (explicit) which gives similar results to the 
Margins Sorted Approval method I now support.  And others are possible.

Chris B.

On 17/05/2024 10:55 am, Michael Ossipoff wrote:
>
> Elect the candidate not beaten among all the strongest defeats that 
> don’t contradict (form any cycles with) eachother.
> ————-
> In my recent Approval-post, I forgot to mention (but have mentioned 
> elsewhere) that Approval’s Myerson-Weber equilibrium is at the 
> voter-median.
>
> i.e. Approval homes-in on where the CW is.
>
> (A Myerson-Weber equilibrium is an outcome consistent with, 
> confirming, the voters’ beliefs that led to that outcome.)
>
> ----
> Election-Methods mailing list - see https://electorama.com/em for list info